The Branch has a long-standing involvement in the systematic evaluation of chemotherapeutic drugs alone or in combination with radiotherapy. As more individuals become long-term cancer survivors, there is greater need to evaluate risk-benefit ratios for various treatment protocols. Branch studies have shown that platinum-based chemotherapy for ovarian cancer was linked to a four-fold risk of leukemia; however, the substantial benefit that platinum-based treatment offers, outweighs the relatively small excess risk of leukemia. A study of leukemia following radiotherapy for testicular cancer revealed a three-fold elevated risk for leukemia, and this risk increased with increasing radiation dose to active bone marrow. The risk of solid tumors among 16,367 patients with chronic lymphocytic leukemia (CLL)was quantified. The overall observed to expected (O/E) ratio was 1.20 (p <0.05). Signficant 2 to 5-fold excess risks were noted for Kaposi sarcoma, malignant melanoma, and cancers of the larynx and lung. Additional investigations of second cancers after CLL are needed to explore the role of immunologic impairment and other etiologic influences. Secondary Lung Cancer following Hodgkin's Disease. Second malignant neoplasms constitute the most common cause of mortality in patients cured of Hodgkin's disease (HD), with lung cancer representing the most frequent solid tumor. Despite the frequency of this cancer, analytic data which describe the importance of chemotherapy and radiotherapy in its development have been conflicting, sparse and inadequately controlled for smoking. REB conducted a nested case-control study of lung cancer within a cohort of 19,046 one-year survivors of HD. Alkylating agents without radiotherapy increased lung cancer risk 4.2-fold (95% CI = 2.1-8.8). Radiation without alkylating agents increased risk 5.9-fold Risk increased with both increasing number of cycles of alkylating agents and increasing radiation dose (P trend < .001). Risk increased with cumulative amounts of mechlorethamine and procarbazine (P < .001) when evaluated separately. Risk after alkylating agents and radiotherapy in combination was as expected, if individual excess risks were summed. Tobacco use increased lung cancer risk 20-fold; risks from smoking appeared to multiply risks from treatment. In a separate molecular study, we found that the overall frequency of microsatellite alterations in lung cancers after HD, was substantially greater (2.4-fold, P =.004) than in sporadic tumors. No differences in the pattern of TP53 and K-ras mutations were detected between post-HD and sporadic lung cancers. Although the mechanisms underlying the development of increased microsatellite alterations are not clear, they have been associated with several factors, including immunosuppression and radiation exposure. Significantly elevated risks of stomach, breast and uterine cervix cancers were observed up to 25 years following diagnosis of Hodgkin's disease in 32,000 patients. The actuarial risk of developing a solid tumor after 25 years was 21.9%. A study focusing on a subset of 6,000 children and adolescents with Hodgkin's disease indicated a high risk for second cancers of the thyroid and respiratory tract in patients treated before age ten; at older ages, the highest risks were seen for cancers of the digestive tract and breast. An evaluation of new cancers in 29,737 allogeneic BMT recipients demonstrated for the first time that established risk factors for early-onset, post-transplant lymphoproliferative disorders do not predict the late-onset of this disorder. Late developing lymphomas occur more frequently following extensive graft-versus-host disease, an established correlate of immune dysregulation in long term survivors. Among 2,739 lymphoma patients undergoing autologous transplants, the risks of leukemia and myelodysplastic syndrome were associated with the intensity of pretransplant chemotherapy. A cohort of 1,600 Retinoblastoma(RB)patients continues to be monitored for cancer risk, and RB patients who have developed melanoma are undergoing clinical examination for dysplastic nevi syndrome, lipomas, and mutations in melanoma susceptibility genes. DNA collected from hereditary patients with a second cancer are being genotyped for mutations in their RB1 gene. A recent Branch survey of RB patients indicated that cigarette smoking was less in hereditary patients compared with sporadic patients, providing further support for a link between a mutation in RB1 gene and the appearance of lung cancer. In collaboration with the Genetic Epidemiology Branch and cancer registries in four Nordic countries, breast cancer risk was reported to be increased in mothers, but not other female relatives, of patients with ataxia-telangiectasia. Expansion of the original cohort and sequencing of the ATM gene in patients and family members is also in progress. A multi-center retrospective cohort study of 5,573 women with scoliosis revealed a statistically significant 1.7-fold risk of death from breast cancer. Patterns were consistent with a radiation etiology in that risk increased with increasing number of diagnostic x-ray examinations and estimated cumulative radiation dose to the breast. Potential confounding between radiation dose and severity of scoliosis may explain some of the excess risk observed. The Branch is currently estimating the risk of cancer after exposure to Thorotrast among several large populations. Thorotrast, once used as an angiographic contrast agent, is not excreted from the body to any appreciable extent; and has induced high rates of liver angiosarcoma and leukemia. The Branch recently completed a 40-year mortality follow-up of 693 Swedish patients with neurological disorders who received Thorotrast. A significant dose-response relation was found for all causes of death combined, and all malignant tumors combined. A second followup study of cancer mortality in a cohort of 2000 irradiated and 2000 surgically treated peptic ulcer patients demonstrated increased risks of stomach, pancreatic and lung cancer following radiotherapy in long-term survivors. Coronary heart disease but not stroke in this population also appear to be linked with radiotherapy. A retrospective cohort study of cancer mortality in 5,300 subjects treated with nasopharyngeal irradiation (radium) during childhood and 5,200 control subjects in the Netherlands revealed an excess risk of non-Hodgkin's lymphoma. Overall, the risk of head and neck cancers was not increased. The results suggested excess thyroid cancer risk based on small numbers, but no evidence of highly increased risk of non-melanoma skin cancer, benign head and neck tumors or self-reported hormonal disorders. A cohort of female tuberculosis patients continues to be studied for the risk of breast cancer following multiple chest fluorscopies received during adolescence and early adult life. Previous studies indicated a dose-response relationship between number of fluorscopies and breast cancer. Preparation of a monograph describing multiple primary cancers in SEER (1973-1998)is underway. The purpose is to clarify the patterns of multiple primary cancers in SEER through a comprehensive evaluation of cancer risk following each initial primary cancer, and to integrate these findings with published literature to further our understanding of cancer etiology. Nested case-control studies of brain, breast and thyroid cancers following an initial childhood cancer are being planned within the large cohort of 5-year childhood cancer survivors. Studies will focus on treatment-induced second cancers. Detailed radiation dosimetry is in progress as well as methodologic work regarding control selection and issues of overmatching. A special issue of Radiation Research describing dosimetry methods that have been developed to estimate and reconstruct medical, environmental and occupational radiation exposures in epidemiologic studies conducted by the Branch is in preparation. Pediatric CT scans and radiation risk were described in a newsletter prepared in collaboration with the Society for Pediatric Radiology. The purpose was to inform physicians of the increased risk from CT radiation and strategies to lower the dose that children may receive. Several studies of thyroid cancer etiology have been conducted. Pooled analyses of 12 international case-control studies of thyroid cancer concluded that consumption of fish did not increase thyroid cancer risk, and cruciferous vegetables, which contain goitrogenic substances as well as several constituents that can inhibit carcinogenesis, were weakly related to a reduced risk of thyroid cancer. Medullary thyroid cancer was associated with history of thyroid nodules, hypertension, gallbladder disease and allergies. A cohort of children irradiated in childhood for enlarged tonsils continues to be followed for the occurrence of thyroid and other head and neck cancers. A study investigating thyroid cancer risk following diagnostic x-rays is comparing medical record data with questionnaire data to evaluate recall bias. There was some indication of relatively poorer reporting among controls for certain types of xrays and large number of xrays. However, estimates of the risk associated with the exposure to diagnostic xrays were similar, regardless of whether interview or medical record data were used. Methodologic studies are evaluating dose uncertainty in a study of radiation-induced thyroid cancer in a cohort of Israeli children treated for tinea capitis. The Branch is also evaluating the risk of cancer following diagnostic and therapeutic Iodine-131 in several large, international study populations. Cancer mortality following radiation treatment for benign gynecologic disease in a cohort of 12,000 women is being updated to evaluate solid tumor risk. An analysis of second cancers occurring among 1,498 male breast cancer patients in the SEER cancer registry program found a large increased risk of developing a contralateral breast cancer (9 cases, SIR = 30). The risk for contralateral breast cancer was especially high for men diagnosed with their first cancer before age 50 years. Analysis of BRCA1/ 2 in Israeli male breast cancer tumor tissue revealed three founder mutations in 122 breast cancer tissue samples; 16 cancers had 6174delT mutations and 4 had 185delAG mutations. Mutations were found in both Ashkenazi and non-Ashkenazi Jews.